We model a decision impact pathway is for school gardens as a general intervention for sustainable children’s food environments in urban Hanoi, Vietnam (Whitney et al. 2024).
Conceptual model of school gardens as an intervention. Should urban Hanoi school boards invest time and money in creating school gardens? Should they invest in formal STEM education as part of these gardens?
Simulation of the school garden intervention options:
# Source our model
source("CODAS_Garden_Model.R")
# Ensure consistent results with the random number generator
# not for each 'run' of the MC simulation but for
# consistency each time we run the entire simulation
set.seed(1234)
garden_simulation_results <- mcSimulation(
estimate = estimate_read_csv("data/inputs_school_garden.csv"),
model_function = school_garden_function,
numberOfModelRuns = 1e4, #run 10,000 times
functionSyntax = "plainNames"
)
The Net Present Value (i.e. current value of the future benefits) of the garden decision options over 5 years of the intervention. For public and private schools the STEM costs are considered to be in the same garden space but with the additional costs and benefits of a full STEM education program. All options are compared to the same years of using the land for something that is not related to the garden, i.e. as a playground or for parking. Here we plot the distribution for the decision and frame the projected NPV.
source("functions/plot_distributions.R")
plot_distributions(mcSimulation_object = garden_simulation_results,
vars = c("NPV_garden","NPV_garden_STEM",
"NPV_garden_public_school",
"NPV_garden_STEM_public_school"),
method = 'smooth_simple_overlay',
base_size = 7,
x_axis_name = "Comparative NPV outcomes")
The same results as boxplots:
source("functions/plot_distributions.R")
plot_distributions(mcSimulation_object = garden_simulation_results,
vars = c("NPV_garden","NPV_garden_STEM", "NPV_garden_public_school", "NPV_garden_STEM_public_school"),
method = "boxplot",
base_size = 7,
x_axis_name = "Comparative NPV outcomes")
summary(garden_simulation_results$y$NPV_garden)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -1394.3 -159.6 410.8 706.2 1247.5 10467.9
Summary of the savings for the formal STEM education garden option
summary(garden_simulation_results$y$NPV_garden_STEM)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -4000.8 -481.3 139.1 354.1 978.1 10142.9
Total expected costs for a school garden
summary(garden_simulation_results$y$total_costs)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 76.52 200.89 434.57 399.79 516.82 1604.26
Total expected costs for a school garden with STEM education
summary(garden_simulation_results$y$total_costs_STEM)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 137.3 365.1 853.8 938.1 1263.6 5272.5
First year expected costs for a school garden
summary(garden_simulation_results$y$Cashflow_garden1)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -664.22 -93.99 14.41 68.96 179.07 2042.84
First year expected costs for a school garden with STEM education
summary(garden_simulation_results$y$Cashflow_garden_STEM1)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -1003.78 -237.64 -119.10 -76.55 47.19 1933.97
Here we assess value of information with the multi_EVPI
function.
# Subset the outputs from the mcSimulation function (y) by selecting the correct variables be sure to run the multi_EVPI only on the variables that we want. Find them with names(garden_simulation_results$y)
mcSimulation_table <- data.frame(garden_simulation_results$x,
garden_simulation_results$y[1:9])
Value of information for the garden option (no STEM)
source("functions/multi_EVPI.R")
# first_out_var is the first result variable in the table, "NPV_garden" in our case.
# names(garden_simulation_results$y)
# evpi <- multi_EVPI(mc = mcSimulation_table, first_out_var = "NPV_garden")
# save as a local .csv (takes ~ a day to run this)
# save(evpi,file="data/data_evpi.Rda")
load("data/data_evpi.Rda")
# open from saved file (last model run)
#Value of information the garden intervention decision
source("functions/plot_evpi.R")
plot_evpi_garden <- plot_evpi(evpi, decision_vars = "NPV_garden")
# In this case there were no high value variables. We can be confident in supporting the decision, even with our imperfect current understanding.
# Value of information for the garden option with formal STEM education.
# using the results of the same multi_EVPI
plot_evpi_STEM <- plot_evpi(evpi, decision_vars = "NPV_garden_STEM")
# Value of information for the public school garden option with no formal STEM education.
# using the results of the same multi_EVPI
plot_evpi_public <- plot_evpi(evpi, decision_vars = "NPV_garden_public_school")
# Value of information for the public school garden option with formal STEM education.
# using the results of the same multi_EVPI
plot_evpi_public_STEM <- plot_evpi(evpi, decision_vars = "NPV_garden_STEM_public_school")
# Cashflow of the garden option without formal STEM education
# This will be the cost for public and private schools over the intervention.
source("functions/plot_cashflow.R")
plot_cashflow_garden <- plot_cashflow(mcSimulation_object = garden_simulation_results,
cashflow_var_name = "Cashflow_garden")
# Cashflow of the garden option with formal STEM education
source("functions/plot_cashflow.R")
plot_cashflow_STEM <- plot_cashflow(mcSimulation_object = garden_simulation_results,
cashflow_var_name = "Cashflow_garden_STEM")
# Projection to Latent Structures (PLS)
# We use Projection to Latent Structures (PLS) model to get some sense of the correlation strength and direction for model variables and our outcome variables.
# For passive education garden option
source("functions/pls_model.R")
pls_result <- pls_model(object = garden_simulation_results,
resultName = names(garden_simulation_results$y)[1], # the "NPV_garden"
ncomp = 1)
# read in the common input table
input_table <- read.csv("data/inputs_school_garden.csv")
# source the plot function
source("functions/plot_pls.R")
plot_pls_garden <- plot_pls(pls_result,
input_table = input_table, threshold = 0.9)
#For school garden with formal STEM education
pls_result_STEM <- pls_model(object = garden_simulation_results,
resultName = names(garden_simulation_results$y)[2], # the "NPV_garden_STEM"
ncomp = 1)
plot_pls_STEM <- plot_pls(pls_result_STEM, input_table = input_table, threshold = 0.9)
#plot NPV as boxplot, pls, and evpi, evpi share names with pls
#each to a row
#shared x axis (values) and y axis (names)
dist_plot_NPV_garden_public <- plot_distributions(
mcSimulation_object = garden_simulation_results,
vars = "NPV_garden_public_school",
method = "boxplot",
base_size = 7)
dist_plot_NPV_garden_STEM_public <- plot_distributions(
mcSimulation_object = garden_simulation_results,
vars = "NPV_garden_STEM_public_school",
method = "boxplot",
base_size = 7)
# install.packages("devtools")
# devtools::install_github("thomasp85/patchwork")
library(patchwork)
dist_plot_NPV_garden_public | dist_plot_NPV_garden_STEM_public /
plot_pls_garden | plot_pls_STEM /
plot_evpi_STEM | plot_evpi_garden
#plot NPV as boxplot, pls, and evpi, evpi share names with pls
#each to a row
#shared x axis (values) and y axis (names)
dist_plot_NPV_garden <- plot_distributions(
mcSimulation_object = garden_simulation_results,
vars = "NPV_garden",
method = "boxplot",
base_size = 7)
dist_plot_NPV_garden_STEM <- plot_distributions(
mcSimulation_object = garden_simulation_results,
vars = "NPV_garden_STEM",
method = "boxplot",
base_size = 7)
# "NPV_garden_public_school", "NPV_garden_STEM_public_school"),
# install.packages("devtools")
# devtools::install_github("thomasp85/patchwork")
library(patchwork)
(dist_plot_NPV_garden+ inset_element(plot_cashflow_garden, 0.6, 0.6, 1, 1) | plot_pls_garden | plot_evpi_garden) /
(dist_plot_NPV_garden_STEM+ inset_element(plot_cashflow_STEM, 0.6, 0.6, 1, 1) | plot_pls_STEM | plot_evpi_STEM)
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning: Removed 15 rows containing missing values or values outside the scale range
## (`geom_line()`).
## Removed 15 rows containing missing values or values outside the scale range
## (`geom_line()`).
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_line()`).
## Removed 5 rows containing missing values or values outside the scale range
## (`geom_line()`).
Cashflow for public and private schools
plot_cashflow_garden | plot_cashflow_STEM
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning: Removed 15 rows containing missing values or values outside the scale range
## (`geom_line()`).
## Removed 15 rows containing missing values or values outside the scale range
## (`geom_line()`).
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning in FUN(X[[i]], ...): NAs introduced by coercion
## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_line()`).
## Removed 5 rows containing missing values or values outside the scale range
## (`geom_line()`).
source("functions/plot_pareto.R")
private_paretto <- paretto_front(
economic_return_garden = garden_simulation_results$y$NPV_garden,
child_health_garden = garden_simulation_results$y$health,
biodiversity_garden = garden_simulation_results$y$biodiversity,
economic_return_STEM = garden_simulation_results$y$NPV_garden_STEM,
child_health_STEM = garden_simulation_results$y$health_STEM,
biodiversity_STEM = garden_simulation_results$y$biodiversity
)
ggplotly(private_paretto)
source("functions/plot_pareto.R")
public_paretto <- paretto_front(
economic_return_garden = garden_simulation_results$y$NPV_garden_public_school,
child_health_garden = garden_simulation_results$y$health,
biodiversity_garden = garden_simulation_results$y$biodiversity,
economic_return_STEM = garden_simulation_results$y$NPV_garden_STEM_public_school,
child_health_STEM = garden_simulation_results$y$health_STEM,
biodiversity_STEM = garden_simulation_results$y$biodiversity
)
ggplotly(public_paretto)
Here we provide a summary of the garden intervention options. We do
this with a summary table of the simulation results. We show the
percentage of missing values as well as the mean, median and standard
deviation (SD) for each output of our model simulations. We use the
gt_plt_summary() from {gtExtras} and with options from
{svglite}. The table shows the name, the plot overview as well as the
number of missing values, the mean, median and the standard deviation of
the distribution for all variables that were fed into the model from our
input table of uncertainty values.
# Subset the outputs from the mcSimulation function (y) to summarize only on the variables that we want.
# names(garden_simulation_results$x)
mcSimulation_table_x <- data.frame(garden_simulation_results$x[4:7,21:30]) #,
# names(garden_simulation_results$y)
gtExtras::gt_plt_summary(mcSimulation_table_x)
| mcSimulation_table_x | ||||||
| 4 rows x 10 cols | ||||||
| Column | Plot Overview | Missing | Mean | Median | SD | |
|---|---|---|---|---|---|---|
| equipment_cost | 0.0% | 9.6 | 9.3 | 2.6 | ||
| construction_cost | 0.0% | 57.0 | 35.7 | 50.3 | ||
| garden_designing_costs | 0.0% | 12.7 | 12.4 | 1.2 | ||
| teacher_training_cost | 0.0% | 12.6 | 12.1 | 3.7 | ||
| school_board_planning | 0.0% | 8.6 | 8.5 | 0.9 | ||
| teaching_equipment | 0.0% | 7.9 | 7.9 | 0.5 | ||
| compost_starting | 0.0% | 7.3 | 6.9 | 1.2 | ||
| worm_starting | 0.0% | 5.4 | 5.1 | 2.6 | ||
| livestock_establishment_costs | 0.0% | 15.8 | 16.7 | 6.2 | ||
| fishpond_cost | 0.0% | 8.0 | 7.9 | 0.8 | ||
# a summary table with missing, mean, median and sd
The full repository can be accessed at https://github.com/CWWhitney/urban_school_gardens